ThinkMind // International Journal On Advances in Intelligent Systems, volume 3, numbers 3 and 4, 2010 // View article intsys_v3_n34_2010_12
Policies and Abductive Logic: An Approach to Diagnosis in Autonomic Management
Keywords: Autonomic Computing; Policy; Policy-based Management; Diagnosis; Abduction
Policy-based Autonomic Management monitors a system and its applications and tweaks performance parameters in real-time based on a set of governing policies. A policy specifies a set of conditions under which one or more of a set of actions are to be performed. It is very common that multiple policies' conditions are met simultaneously, each advocating many actions. Deciding which action to perform is a non-trivial task. We propose a method of diagnosing the system to try to determine the best action or actions to perform in a given situation using Abductive Inference. We develop an original method of building a causality graph to facilitate diagnosis directly from a set of policies. We propose two alternate methods of ranking diagnosis hypotheses based on their likelihood of success. Performance of the diagnosis method is evaluated within an autonomic management system monitoring the performance of a LAMP (Linux, Apache, MySQL, PHP) server being governed by the manager. The performance of the diagnosis method is compared against previous methods used by an existing autonomic manager. The results are favourable when compared to previous methods of action selection and to the server running without the autonomic manager. A walkthrough of an example experiment using diagnosis is presented to gain additional insight into the method.
Pages: 313 to 325
Copyright: Copyright (c) to authors, 2010. Used with permission.
Publication date: April 6, 2011
Published in: journal